RAG & retrieval skills for AI agents

RAG & retrieval

912 rows from 257 repositories

Putting the right documents in front of the model.

What RAG & retrieval skills agree on

229 skill files read, by 199 of the 213 authors on this shelf whose files we hold, 2026-09-06

The middle one of the 39 measured here is ~1.8k tokens long, counted with cl100k_base

Counted by distinct author, so one author publishing three of these counts once. Where a claim sits in fewer files than authors, that is said: a claim held by forty authors across three files is one file people copied, not forty people who agreed. Near-identical wordings are grouped and the other wordings are shown, so the grouping is yours to check.

What they tell the agent to do

Enable caching for frequent queries14 of 199 in 5 filesalso worded as Cache frequent queries
Enable MMR for diverse results12 of 199 in 5 filesalso worded as Use MMR for diverse results; Diversify results using MMR
Enable binary quantization to reduce memory11 of 199 in 4 filesalso worded as use binary quantization to reduce memory; Enable quantization to reduce memory
Initialize the database with dimensions matching the embedding model11 of 199 in 4 filesalso worded as Match database dimensions to the embedding model
Start the similarity threshold at 0.711 of 199 in 4 filesalso worded as Start similarity threshold at 0.7
Check database statistics when diagnosing slow search11 of 199 in 4 filesalso worded as Check database statistics when search is slow; run stats to diagnose slow searches
Export and import vectors as JSON10 of 199 in 3 filesalso worded as Export vectors to JSON for backup; export vectors to JSON for backups
Match index dimension to the embedding model10 of 199 in 9 filesalso worded as Match index dimensions to the embedding model output; match query and index embedding dimensions
Combine vector similarity with metadata filters9 of 199 in 3 filesalso worded as Combine vector similarity with metadata filtering; Combine vector similarity with conventional where filters
Measure precision, recall, MRR, and NDCG9 of 199 in 7 filesalso worded as Measure precision@k, recall@k, MRR, and NDCG; Calculate precision@k, recall@k, MRR, or nDCG
Order filters cheap before expensive9 of 199 in 2 files
Generate a runnable scaffold in the user's stack9 of 199 in 2 files
Recommend multi-action scoring when frequent tuning is expected9 of 199 in 2 files
Batch store documents for bulk inserts9 of 199 in 2 filesalso worded as batch insert vectors for bulk storage
List hydrations required by filters and scorers9 of 199 in 2 filesalso worded as List hydrations each filter and scorer needs

What they tell it not to do

Do not mix embedding models10 of 199 in 7 filesalso worded as Don't mix embeddings of different dimensions; Never mix embedding models
Do not skip metadata enrichment9 of 199 in 7 filesalso worded as Do not skip metadata enrichment; Skip metadata enrichment
Do not invent benchmark numbers9 of 199 in 2 filesalso worded as Do not invent benchmark numbers
Do not use 'For You' branding or 'X-like' naming9 of 199 in 2 filesalso worded as Do not use 'For You' branding or 'X-like' naming
Do not let side effects block the response9 of 199 in 2 filesalso worded as Do not let side effects block the response
Do not score before filtering9 of 199 in 2 filesalso worded as Do not score before filtering
Do not deploy without citation validation tests8 of 199 in 6 filesalso worded as Do not deploy without citation validation tests; Do not deploy without testing on production-like data volumes
Never use default chunk size 512 without domain evaluation8 of 199 in 6 filesalso worded as Never use default chunk size 512 without domain evaluation; Do not use default chunk size without evaluation
Never commit .env files8 of 199 in 5 filesalso worded as Do not commit env files; Never commit .env
Do not couple the embedding model tightly to application code7 of 199 in 5 filesalso worded as Do not couple the embedding model tightly to application code; Do not couple embedding model tightly to application code

What they expect to be installed

Pinecone22 of 199 in 19 files
pgvector20 of 199 in 17 files
OpenAI embeddings API17 of 199 in 12 files
Qdrant16 of 199
RAGAS16 of 199 in 14 files
curl15 of 199
sentence-transformers15 of 199 in 11 files
ChromaDB13 of 199 in 12 files
Weaviate13 of 199 in 10 files
pip12 of 199 in 11 files

What they ask it to produce

Initialized vector database file11 of 199 in 4 filesalso worded as vector database file
JSON-formatted query results10 of 199 in 3 filesalso worded as JSON query results
JSON export of vectors9 of 199 in 3 files
Runnable pipeline scaffold in the user's stack9 of 199 in 2 files
Retrieval pipeline design showing query-to-results flow7 of 199 in 5 filesalso worded as Retrieval pipeline design; Retrieval pipeline design from query to results
Vector database selection with trade-off analysis6 of 199 in 4 files
Chunking strategy with examples and rationale6 of 199 in 4 files
Actionable steps with verification6 of 199 in 2 files

When RAG & retrieval authors say to reach for one

The situations these authors wrote into their own files, counted out of the same 199 authors, with the skills that name each one

Building RAG systems28 of 199 in 16 files
- Rag architect
- Embedding optimization
- Agentdb vector search
- and 7 more on this shelf
Building semantic search engines18 of 199 in 10 files
- Agentdb vector search
- Ai searching docs
- and 7 more on this shelf
Building intelligent knowledge bases12 of 199 in 5 files
- Agentdb vector search
- and 7 more on this shelf
Implementing semantic search11 of 199
Building a system that picks top K items for a user9 of 199 in 2 files
- Recsys pipeline architect
- and 4 more on this shelf
User asks how to rank something or describes a feed problem9 of 199 in 2 files
- Recsys pipeline architect
- and 4 more on this shelf
User has a scoring function needing pipeline plumbing9 of 199 in 2 files
- Recsys pipeline architect
- and 4 more on this shelf
User mentions recommendation, feed, or ranking keywords9 of 199 in 2 files
- Recsys pipeline architect
- and 4 more on this shelf

How RAG & retrieval skills are built

822 skill directories by 226 authors, read from their repositories’ own file trees 2026-08-05

The middle bundle among those shipping files is 2 files, 16.5 KB beside SKILL.md

Counted by distinct author, same as above, so one author publishing forty template copies counts once. SKILL.md itself is not counted as a file, so a single-file skill is one where that file is the whole skill.

The shape

SKILL.md is the whole skill99 of 226 authors, 494 of 822 skills
files ship beside it127 of 226 authors, 328 of 822 skills
executable scripts ship inside41 of 226 authors, 81 of 822 skills

The folders they converge on

references/78 of 226 authors, 180 of 822 skills
scripts/31 of 226 authors, 65 of 822 skills
templates/10 of 226 authors, 14 of 822 skills
examples/9 of 226 authors, 22 of 822 skills
evals/7 of 226 authors, 12 of 822 skills
agents/6 of 226 authors, 9 of 822 skills

Haiku rag
ggozad/haiku.rag/io.github.ggozad/haiku-rag MCP server
567★ repo
Opinionated agentic RAG powered by LanceDB, Pydantic AI, and Docling
Mcp local rag
shinpr/mcp-local-rag/io.github.shinpr/mcp-local-rag MCP server
360★ repo
Easy-to-setup local RAG server with minimal configuration
Chunking strategy
giuseppe-trisciuoglio/developer-kit/plugins/developer-kit-ai/skills/chunking-strategy Skill
329★ repo
Modular plugin marketplace for Claude Code and agentic CLIs, with validated, spec-driven skills, agents, commands, and workflows for Java, TypeScript, Python, PHP, AWS, and AI.
Qdrant advisor
qdrant/skills/meta/qdrant-advisor Skill
220★ repo
Agent skills for Qdrant vector search: scaling, performance optimization, search quality, monitoring, deployment, model migration, version upgrades, and SDK usage across Python, TypeScript, Rust, Go, .NET, Java
Mcp server
BingoWon/apple-rag-mcp/com.apple-rag/mcp-server MCP server
115★ repo
Apple Developer Documentation with Semantic Search, RAG, and AI reranking for MCP clients
OmniClip RAG
msjsc001/OmniClip-RAG/io.github.msjsc001/omniclip-rag-mcp MCP server
45★ repo
Read-only local-first MCP server for private Markdown, PDF, and Tika-backed search on Windows.
Local FAISS MCP Server
nonatofabio/local_faiss_mcp/io.github.nonatofabio/local-faiss-mcp MCP server
33★ repo
Local FAISS vector database for RAG with document ingestion, semantic search, and MCP prompts.
Coff0xc ai agent rag
Coff0xc/coffee-skill/skills/coff0xc-ai-agent-rag Skill
no license15★ repo
Installable Codex/AgentSkills workflow pack for dev, Agent/RAG, API/data, Office artifacts, research diagrams, and authorized security review.
Pinecone assistant
pinecone-io/skills/skills/pinecone-assistant Skill
14★ repo
Pinecone's official Agent Skills library, for use with agentic IDEs such as Cursor, Github Copilot, Antigravity, Gemini CLI and more.
Mcp local rag
damoqiongqiu/mcp-local-rag/io.github.damoqiongqiu/mcp-local-rag MCP server
13★ repo
Semantic code & doc search with keyword boost. AST code nav, auto HF mirror, local, privacy-first.
Zvec
zvec-ai/zvec-agent-skills/skills/zvec Skill
12★ repo
Official AI Agent skills for building with Zvec vector database
Powabase
powabase-ai/agent-skills/skills/powabase Skill
9★ repo
Agent Skills that help AI coding assistants build on Powabase — the AI Backend-as-a-Service (RAG, agents, orchestration, workflows) plus a Supabase-style BaaS layer.
Catsu
feyninc/skills/skills/catsu Skill
6★ repo
agent-skills for chonkie's ecosystem
Wso2 docs mcp server
iamvirul/wso2-docs-mcp-server/io.github.iamvirul/wso2-docs-mcp-server MCP server
3★ repo
Semantic search over WSO2 docs (APIM, MI, Choreo, Ballerina) via RAG and pgvector.
Ai ml
iuliandita/skills/skills/ai-ml Skill
2★ repo
46 skills for DevOps, security, infra, and software engineering, maintained with lint/spec checks, behavioral test coverage, and a Karpathy-style autoresearch loop.
Rag query
butchokoy25/lightrag-claude-skills/skills/rag-query Skill
2★ repo
Persistent memory for Claude Code via LightRAG knowledge graphs — 7 skills + 2 auto-hooks
Calypso Multimodal RAG MCP
calypso-so/calypso-mcp-server/io.github.calypso-so/multimodal-rag-mcp-server MCP server
2★ repo
Calypso multimodal RAG for grounded answers from docs, images, charts, and knowledge.
Knowledge base rag
igorbeethetech/skills/skills/knowledge-base-rag Skill
1★ repo
Public repo for skills
Pageindex rag
mmtmr/pageindex-rag Skill
no license1★
PageIndex RAG: Reasoning-based retrieval architecture replacing vector databases with hierarchical navigation
Food recipe mcp
AIDataNordic/Food-Recipe-MCP/io.github.AIDataNordic/food-recipe-mcp MCP server
1★ repo
Semantic search across 50,000+ food recipes with hybrid retrieval and reranking.
Rag vault
RobThePCGuy/rag-vault/io.github.RobThePCGuy/rag-vault MCP server
no license1★ repo
Local RAG MCP server with hybrid search, PDF/DOCX support, and zero-config setup
TheCrawler
manchittlab/TheCrawler/io.github.manchittlab/thecrawler MCP server
1★ repo
Universal web scraper with LLM-ready markdown, RAG chunking, PDF/DOCX support.
Postgres first
Thanane15M/postgres-first Skill
0★
Replace Redis, MongoDB, n8n, Elasticsearch with native PostgreSQL. A Claude Skill.
Boardgame rag bot
mike623/agent-skills/skills/boardgame-rag-bot Skill
no license0★ repo
Eval harness architect
satishTheLegend/eval-harness-architect Skill
0★
Stand up a rigorous, regression-proof evaluation harness for any LLM/agent system from zero — datasets, scorers, CI gates, and drift monitoring.
Goal chunking agent architect
halans/goal-chunking-agent-architect Skill
no license0★
An Agent Skill that teaches your coding agent to design other (sub)agents well.
Azpg ai in database
lestermarch/postgres-ai-experts/skills/azpg-ai-in-database Skill
no license0★ repo
Composable AI agents and skills for operating Azure Database for PostgreSQL Flexible Server - PostgreSQL can be used for everything.
Embgrep
QuartzUnit/embgrep/io.github.ArkNill/embgrep MCP server
0★ repo
Local semantic search — embedding-powered grep for files, zero external services.
Rag knowledge graph mcp
CSOAI-ORG/rag-knowledge-graph-mcp/io.github.CSOAI-ORG/rag-knowledge-graph-mcp MCP server
0★ repo
rag-knowledge-graph-mcp MCP server by MEOK AI Labs
Rag knowledge mcp
CSOAI-ORG/rag-knowledge-mcp/io.github.CSOAI-ORG/rag-knowledge-mcp MCP server
0★ repo
rag-knowledge-mcp MCP server by MEOK AI Labs
Obsidian Semantic Search
DalecB/obsidian-semantic-mcp/io.github.DalecB/obsidian-semantic-mcp MCP server
0★ repo
Read-only semantic search MCP server for Obsidian vaults using local Ollama embeddings.
Mcp file processor
IntelagentStudios/Intelagent-MCPs/io.github.IntelagentStudios/mcp-file-processor MCP server
0★ repo
Text extraction, keyword extraction, language detection, and chunking for RAG
Deeprepo
abhishek2432001/deeprepo/io.github.abhishek2432001/deeprepo MCP server
0★ repo
Productivity-boosting RAG engine for codebases with multi-provider AI support and semantic search.
Rag
fieldcure/fieldcure-mcp-rag/io.github.fieldcure/rag MCP server
0★ repo
MCP RAG server with hybrid search, multi-KB support, and AI-powered chunk contextualization.
Rag mcp
jaimenbell/rag-mcp/io.github.jaimenbell/rag-mcp MCP server
0★ repo
Minimal RAG-over-a-corpus MCP retrieval: search_knowledge returns cited chunks. Local embeddings.
Enterprise Internal Knowledge Base: Production-Ready RAG + MCP
kimsb2429/internal-knowledge-base/io.github.kimsb2429/internal-knowledge-base MCP server
0★ repo
Production-ready RAG + MCP demo: eval-in-CI merge gate, Langfuse traces, structure-aware chunking.
Embedding search
lazymac2x/embedding-search-api/io.github.lazymac2x/embedding-search MCP server
no license0★ repo
Cloudflare Workers MCP server: embedding-search
Linggen - Local RAG
linggen/linggen-releases/io.github.linggen/linggen MCP server
0★ repo
Local, privacy-focused RAG service for code search via MCP. https://linggen.dev
X Archive RAG
mameshivaa/x-archive-rag/io.github.mameshivaa/x-archive-rag MCP server
0★ repo
Local-first MCP tools for searching and drafting from your X/Twitter archive.
Unstructured Transform
Unstructured-IO/unstructured-mcp-integrations/io.unstructured/transform MCP server
0★ repo
Turn documents into structured, AI-ready data by parsing, enriching, chunking, and embedding.
Ai data engineering
ancoleman/ai-design-components/skills/ai-data-engineering Skill
503★ repo
Comprehensive UI/UX and Backend component design skills for AI-assisted development with Claude
Qdrant clients sdk
qdrant/skills/skills/qdrant-clients-sdk Skill
220★ repo
Agent skills for Qdrant vector search: scaling, performance optimization, search quality, monitoring, deployment, model migration, version upgrades, and SDK usage across Python, TypeScript, Rust, Go, .NET, Java
Scaffold project
eldermoraes/quarkus-agentic-scaffolding/skills/scaffold-project Skill
27★ repo
Scaffolding skill + always-on conventions (CLAUDE.md/AGENTS.md) for Quarkus + LangChain4j agentic AI apps: AI services, multi-agent workflows, and RAG. Installable in Claude Code, Codex, Copilot, Cursor, and any Agent Skills-compatible agent.
Langchain components
krzysztofsurdy/code-virtuoso/skills/frameworks/langchain/langchain-components Skill
20★ repo
Skills, sub-agents, and playbooks for Claude Code, Cursor, and any Agent Skills-compatible AI coding assistant.
Pinecone cli
pinecone-io/skills/skills/pinecone-cli Skill
14★ repo
Pinecone's official Agent Skills library, for use with agentic IDEs such as Cursor, Github Copilot, Antigravity, Gemini CLI and more.
Zvec
zvec-ai/zvec-agent-skills/src/zvec Skill
12★ repo
Official AI Agent skills for building with Zvec vector database
Semantic search
ahmedibrahim085/Claude-Multi-Agent-Research-System-Skill/.claude/skills/semantic-search Skill
11★ repo
A Multi-Agent Research System using Claude Code Skills. it is inspired by Anthropic claude-agent-sdk-demos
Embedding service
Lin-A1/skills-agent/services/embedding_service Skill
no license7★ repo
根据agent skill理念构建的通用智能体框架
Dify rag pm
samzong/agent-brains/agents/dify-rag-pm/skills/dify-rag-pm Skill
6★ repo
Personal agent brain catalog for coding agents: AGENTS.md, SOUL.md, MEMORY.md, skills, workflows, and a bundled loader.
Chonkie
feyninc/skills/skills/chonkie Skill
6★ repo
agent-skills for chonkie's ecosystem
Cookbook
DjordjeMojsilovic/Djordjes-Claude-Setup/cookbook Skill
no license4★ repo
My personal Claude setup – 11 custom skills, CLAUDE.md config and a complete prompting guide for vibe coding and multi-agent workflows.
Gemini embeddings
akrindev/google-studio-skills/skills/gemini-embeddings Skill
no license4★ repo
google ai studio agent skills
Rag eval harness
surpradhan/claude-code-for-ai-engineers/skills/rag-eval-harness Skill
2★ repo
Open-source preview of *Claude Code for AI Engineers* - a methodology-first skill pack for RAG eval, agent debugging, MCP servers, paper reproduction, and benchmark reporting. Full pack (6 skills, 3 templates, 5 slash commands) at surpradhan.gumroad.com/l/claude-code-for-ai-engineers
Pinecone
phucbm/skills/skills/ai/pinecone Skill
no license2★ repo
Claude Plugin - Personal knowledge base for Claude Code — patterns and integrations across projects
Rag remember
butchokoy25/lightrag-claude-skills/skills/rag-remember Skill
2★ repo
Persistent memory for Claude Code via LightRAG knowledge graphs — 7 skills + 2 auto-hooks
Ai readiness
Nordic-AI/production-readiness-skills/skills/ai-readiness Skill
1★ repo
EU-first, stack-agnostic Claude Skills for auditing and remediating production-readiness across security, compliance, testing, reliability, observability, supply chain, data protection, and scalability.
Ai rag eval guardrails
authenticfake/clike/extensions/vscode/templates/harper-init/.clike/skills/ai-rag-eval-guardrails Skill
no license1★ repo
Clike is an AI-native developer experience layer AI Native Pipeline principles and from some visionar on internet
Ai context engine
ebarti/skills/ai-context-engine Skill
0★ repo
📚 Agent skills distilled from technical books — AI Engineering, Context Engineering, Designing Data-Intensive Applications, and more. Agent-agnostic, plain Markdown. Give your AI agent a bookshelf.
Rag knowledge base
Drvivek34/Skill-Bazaar/ai-ml/rag-knowledge-base Skill
no license0★ repo
🛒 Skill Bazaar — an all-in-one open collection of AI Agent Skills (SKILL.md), organized by category. Part of the Mega AI Bazaar.
Query dma rag
tinhct/lux-agent/.agents/skills/query_dma_rag Skill
no license0★ repo
A multi-agent system for reliable algorithmic auditing and explainable regulatory compliance.

All 912 in the catalog